Abortive initiation and long RNA synthesis - ACS Publications

Abortive initiation and long RNA synthesis. Lianna M. ... David R. Russell , Paul D. Miller , and George N. Bennett .... Purnima Bhargava , George A. ...
0 downloads 0 Views 3MB Size
Biochemistry 1981, 20, 2081-2085

208 1

Abortive Initiation and Long Ribonucleic Acid Synthesist Lianna M. Munson and William S. Reznikoff"

ABSTRACT:

In vitro transcription assays have been used to study the rate of ribonucleic acid (RNA) synthesis from the Escherichia coli lactose promoter mutant lacL8UVS contained on a 203-bp (base pair) restriction fragment. The half-time of long (63-base) R N A production from heparin-resistant RNA polymerase-promoter complexes was found to be related to the amount of oligonucleotides released during the initiation process (abortive initiation). Studies indicate that once a ternary complex between the promoter, RNA polymerase, and

a newly synthesized RNA between seven and nine nucleotides long is formed, abortive initiation is reduced and the rate of synthesis of long RNAs is increased. The promoter for the left inverted repeat of the transposable element Tn5 was also examined. It was observed to have a much slower rate of production of long RNAs, and it released oligonucleotides 4 times as often as the lactose promoter. The correlation between the amount of abortive initiation and the half-time of long RNA production is discussed.

D e s p i t e many years of intensive investigation, the mechanism of transcription initiation remains unclear. Incubation of R N A polymerase with DNA (containing a promoter) at 37 OC results in the formation of a stable promoter-specific complex that is resistant to heparin attack [for a review, see Chamberlin (1976)l. Early experiments performed on T7 bacterial RNA polymerase promoters suggested that once this stable complex was formed addition of nucleoside triphosphates would result in rapid initiation of RNA synthesis (Mangle & Chamberlin, 1974). These results were recently confirmed for T7 promoters (Nierman & Chamberlin, 1979), but very different results were obtained from studies on the lacL8UV5' promoter (Stefan0 & Gralla, 1979). The rate of production of long RNAs from a prebound heparin-resistant complex was found to proceed with a half-time of 1 min. This is 1-2 orders of magnitude slower than that estimated for T7 promoters. These results have stimulated speculation as to whether promoter sequences may affect gene expression through differential rates of initiation of RNA synthesis. DNA-RNA polymerase complexes had previously been observed to undergo an abortive initiation reaction involving the release of short oligonucleotides even in the presence of all four nucleoside triphosphates (McClure & Cech, 1978). Further studies on the facL8UV5 promoter indicated that small oligonucleotidesare synthesized in large quantities from prebound complexes (Carpousis & Gralla, 1980). A recycling model for initiation was proposed to account for the high yield oligonucleotides compared to long RNA transcripts. This model suggests that after each phosphodiester bond is formed a finite probability of dissociating the product exists (for the first 7-9 steps). If the product is dissociated, RNA polymerase can reinitiate RNA synthesis without leaving the DNA template. If product dissociation does not occur, RNA polymerase proceeds on to the next polymerization step. Productive initiation is defined as those initiation events that lead to the synthesis of full-length transcripts (i.e., transcripts that start at the correct site and terminate at or near the end of the restriction fragment). In this paper, we present data that suggest that the recycling of R N A polymerase is responsible for the apparent long

half-time of full-length RNA production. When RNA polymerase is bound in a ternary complex which does not recycle, addition of labeled nucleoside triphosphates leads to rapid synthesis of long RNA products. In an independent system where recycling was increased approximately 4-fold, a similar increase in the half-time of full-length RNA synthesis was observed. The slow production of long RNAs can then be explained as follows. The majority of polymerases are involved in recycling through the initiation process (defined here as the synthesis of the first few phosphodiester bonds). At each step, a certain percentage of the polymerases do not dissociate their RNA product but proceed on to form the next phosphodiester bond. After seven to nine bonds have been formed, elongation continues without further recycling. The percentage of products dissociated at each step &e., the amount of abortive initiation) determines the time it takes to escape this recycling process and synthesize a long RNA. The half-time of long RNA synthesis is thus related to the amount of abortive initiation and is a measure of the rate of escape from the recycling process.

From the Department of Biochemistry, College of Agricultural and Life Sciences, University of Wisconsin, Madison, Wisconsin 53706. Receiued August 6, 1980. Supported by Grant GM19670 from the National Institutes of Health and Grant PCM7910686 from the National Science Foundation. L.M.M. was supported by National Institutes of Health Training Grant GM07215.

0006-2960/8 1/0420-208 1$01.25/0

Materials and Methods Heparin and unlabeled nucleoside triphosphates were obtained from Sigma. [a-32P]UTPwas obtained from Amersham ( J > 350 Ci/mmol). Bis(acryly1cystamine) (BAC) was purchased from Bio-Rad Laboratories. RNA polymerase was a gift from L. Maquat and S. Rothstein. Restriction enzymes were obtained from either New England Biolabs or Bethesda Research Labs Inc. The 203-base-pair (bp) restriction fragment containing the lactose control region was isolated from pRZ3 (wild-type promoter), pRZ3111 (Prill; Maquat et al., 1980), and pRZ3UV5 (LSUV5) by digestion with HaeIII [for plasmid constructions, see Hardies et al. (1 979) and Maquat & Reznikoff (1978)l. The transposon Tn5 promoter was isolated by digestion of pRZ102 (Jorgensen et al., 1979) with HpaI and EcoRI. Restriction fragments were separated on 5% polyacrylamide tube gels containing 0.2% BAC, 10% glycerol, 89 mM Tris-borate, pH 8.3, and 2.8 mM EDTA. The appropriate band was cut out and dissolved in 2-mercaptoethanol as previously described (Hansen, 1976). The dissolved gel was diluted 10-fold with water and applied to a 0.25-mL DEAEI Abbreviations used: EDTA, ethylenediaminetetraacetic acid; Tris, tris(hydroxymethy1)aminomethane; DEAE, diethylaminoethyl; NaDodSO4,sodium dodecyl sulfate; BPB, bromophenol blue; XC, xylene cyanol; lac, lactose operon.

0 198 1 American Chemical Society

2082

MUNSON AND REZNIKOFF

BIOCHEMISTRY

cellulose column (Whatman DE52). The DNA was eluted with 0.01 M Tris, pH 7.9, 1.0 mM EDTA, and 1.0 M NaCl after extensive washing with low-salt buffers. A sample of 3.5 volumes of 100% ethanol was used to precipitate the DNA. The pellet was washed with 95% ethanol, redissolved in 0.01 M Tris, pH 7.9, 0.1 mM EDTA, and 0.01 M NaCI, and dialyzed against the same buffer. Transcription assays were performed in the buffer described by Majors (1975) at 37 OC. A 0.75-pmol aliquot of restriction fragment and 0.75 pmol of RNA polymerase were used per assay, except where noted. Heparin was added to a concentration of 100 pg/mL. When cold nucleoside triphosphates were added in the preincubation experiments, their concentrations were as follows: ATP and GTP, 400 pM; UTP, 20 pM. An additional 200 pM ATP and GTP, 20 pM CTP, IO p M UTP, and 25 pCi of [a-’*P]UTP were added where labeled nucleoside triphosphates were indicated in the preincubation experiments. In assays where there was no preincubation, the UTP concentration in the labeled nucleoside triphosphates was increased to 20 pM. Reaction volumes were either 240 or 80 pL (containing the above-mentioned picomoles of fragment and polymerase) with 30 or I O pL withdrawn per time point. In the former case, the reaction was stopped with 100 pL of a solution containing 30 mM EDTA, 50 pg of tRNA, and 0.3 M sodium acetate. This was then ethanol precipitated and resuspended in sample buffer (20% glycerol, 7 M urea, 0.10% NaDodS04, 89 mM Tris-borate, pH 8.3, BPB, and XC). After being boiled for I min the samples were applied to a 20% polyacrylamide-7 M urea gel and electrophoresed. In the latter case, the reactions were stopped by addition of I O pL of the sample buffer, followed by 1 min a t 100 “C. These were then loaded and electrophoresed as described above. Bands were visualized by autoradiography, excised, and counted by Cerenkov radiation. The relative amount of each oligonucleotide was determined by the number of UMPs it contained and the radioactivity incorporated. A 0.01-pL sample of the reaction mixture was counted by Cerenkov radiation to determine the specific activity of the UTP in the assay. This was used to calculate the picomoles of RNA synthesized. Results Rafe of Producfiue Initiation. Restriction fragments containing the 203-bp lactose control region direct the synthesis of a 63-base RNA transcript (Majors, 1975). For measurement of the rate of synthesis of this transcript independent of the binding process, RNA polymerase is first incubated with the DNA for IO min to allow formation of stable promoterspecific complexes. Heparin is added to inactivate any unbound polymerases, followed by the addition of labeled nucleoside triphosphates. The rate of incorporation of the label into the full-length transcript is then determined (Stefano & Gralla, 1979). We have measured a half-time of productive initiation of approximately I min for the lacL8UV5 promoter. In addition, the wild-type lac promoter and lacP‘1 I 1 promoter were assayed and also found to have a half-time of I min (data not shown). This is in agreement with the half-time obtained by Stefano & Gralla (1979) for the lacLSUV5 promoter. Preincubafion wifh Unlabeled Nucleoside Triphosphates. An assay was developed to determine if the half-time of productive initiation is affected by limited transcription during a preincubation period. RNA polymerase is first incubated for I O min with the restriction fragment containing the promoter. After a I-min heparin challenge, either ATP alone,

“ I

--

0 - 9 a - 6

-

--4

n

-

-2 I

~~

o

3

b

FIOURE 1: Autoradiogram of species produced upon incubation with specificnucleoside triphosphates. The restriction fragment containing the lacL8UVS promoter was used as template. Sizes of RNA produns were determined by eluting RNA from the gel and analyzing by homcchromatography. Markers were a gift from W. McClain. Less than 4% of the RNA in lane b is more than IO nucleotides long. A small percentage of the transcription starts at the G residue located a t -1. This presumably gives rise to the pcntanucleotide in lane a and the decanucleotide in lane b. Reactions contained the following: 400 pM ATP, 20 pM [a-’*P]UTP(lane a); 400 pM ATP, GTP. and 20 p M [a-’zP]UTP (lane b).

ATP and UTP, or ATP, UTP, and G T P are added. Preincubation with the designated unlabeled nucleoside triphosphates is continued for 5 min, then labeled nucleoside triphosphates are added, and the rate of formation of the full-length transcript is measured. The 5’ sequence of lac mRNA is pppApApUpUpGpUpGpApGpC (Maizels, 1973). The addition of only ATP to the transcription assay should allow formation of the dinucleotide pppApA, ATP and UTP should allow R N A polymerase to synthesize the tetranucleotide pppApApUpU, and when ATP, UTP, and G T P are preincubated with the polymerase-DNA complex, RNA producls up lo nine nucleotides should be formed. Figure I shows that these are the major products synthesized during the preincubations. The low level of long transcripts indicates minimal contamination by the omitted nucleoside triphosphates in these experiments. Table I shows the half-times of productive initiation measured after preincubation with the indicated unlabeled nucleoside triphosphates as described above. Preincubation with ATP has no effect on the rate of productive initiation. Preincubation with ATP and UTP has only a minor effect on the half-time, but preincubation with ATP, UTP, and G T P reduced the half-time from 60 s to less than 20 s. Quanfitotion of Oligonucleotides. To further investigate this, we examined all small transcripts synthesized during these experiments. Figure 2 shows an autoradiogram of an assay performed without preincubation with nucleoside triphosphates, as described in the first experiment above. Quantitation of the three major bands (tetramer, hexamer, and full length) yielded an average of I s 1 2 oligonucleotides for each full-length transcript produced. Figure 3A shows the

R A T E OF R N A P R O D U C T I O N

V O L . 20, N O . 8, 1 9 8 1

A B 0 b c d e f g h a b c d e f 9 h - -. -

Table I: Half-times of Productive Initiation" nucleoside triphosphates prescnt during preincubation experiment

ATP -

I

+ + +

2 3 4

VI

UTP -

GTP -

half-time (s)

+

+

45

+

2083

60 60

63